Write a guide for quantizing models using frameworks outside of Compressed-Tensors quantization

[kylesayrs]

Background

• Some users want to quantize models and encode using the compressed-tensors format, but do not necessarily want to use compressed-tensors primitives

Proposed Changes

• Add documentation for encoding a model using compressed tensors format using, for example, QAT

[soyr-redhat]

Hey @kylesayrs I'd be willing to pick this one up, just a few questions before I get started:

1. Where should the guide live? Do we want it in the examples/ subdir, or make a new one for guides/ ?
2. Since compressed-tensors appears to be framework-agnostic at the encoding step, should the guide focus on the ModelCompressor workflow itself rather than writing separate sections per quantization method? If the quantization approach doesn't matter much, we just focus on the encoding step and have a "general" guide?

Please let me know if there's anything I'm missing/ anything that you specifically want from me, and I can get at it!